Abstract
We address the problem of performance analysis and prediction of a class of parallel applications on heterogeneous systems. Our attention is oriented towards workstation networks programmed using message passing libraries. Particularly, we consider a switched Ethernet-based network and we use PVM as parallel tool, adopting the master-worker model with the task farm paradigm.
The simulation applied to the matrix multiplication example yields results in good agreement with the experimental ones.
The model makes possible to estimate the computation and communication times and the idle time due to unbalancing, provided that the computation and communication complexity at the task level is known. In this way we are able to evaluate how the efficiency varies with the task granularity and the degree of heterogeneity of the network.
The analysis can be easily modified to copy with other message passing environments.
This is a preview of subscription content, log in via an institution to check access.
Preview
Unable to display preview. Download preview PDF.
References
Clematis, A., Corana, A.: Modeling performance of heterogeneous parallel computers. Technical Report ICE-CNR n. 3/97 (1997), submitted to Parallel Computing
Clematis, A., Corana, A.: Performance analysis of SPMD algorithms on a network of workstations with virtual shared memory. In: D'Hollander, E., Joubert, G., Peters, F., Trottemberg, U. (eds.): Parallel Computing: Fundamentals, Applications and New Directions. Proc. Parallel Computing 97 (ParCo97). Elsevier (in print)
Corana A.: Computing the correlation dimension on a network of workstations. Concurrency, Practice and Experience (to appear)
Donaldson, V., Berman, F., Paturi, R.: Program speedup in a heterogeneous computing network. J. Parallel and Distributed Computing 21 (1994) 316–322
Schmidt, B.K., Sunderam, V.S.: Empirical analysis of overheads in cluster environments. Concurrency, Practice and Experience 6 (1994) 1–32
Sunderam, V.S., Geist, G.A., Dongarra, J., Manchek, R.: The PVM concurrent computing system: Evolution, experiences, and trends. Parallel Computing 20 (1994) 531–545
Yan, Y., Zhang, X., Song, Y.: An effective and practical performance prediction model for parallel computing on nondedicated heterogeneous NOW. J. Parallel and Distributed Computng 38 (1996) 63–80
Zhang, X., Yan, Y.: Modeling and characterizing parallel computing performance on heterogeneous networks of workstations. In: Proc. Seventh IEEE Symp. Parallel Distributed Processing. IEEE Computer Society Press (1995) 25
Author information
Authors and Affiliations
Editor information
Rights and permissions
Copyright information
© 1998 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Clematis, A., Corana, A. (1998). Performance analysis of task-based algorithms on heterogeneous systems with message passing. In: Alexandrov, V., Dongarra, J. (eds) Recent Advances in Parallel Virtual Machine and Message Passing Interface. EuroPVM/MPI 1998. Lecture Notes in Computer Science, vol 1497. Springer, Berlin, Heidelberg. https://doi.org/10.1007/BFb0056554
Download citation
DOI: https://doi.org/10.1007/BFb0056554
Published:
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-540-65041-6
Online ISBN: 978-3-540-49705-9
eBook Packages: Springer Book Archive